As popular as it is, at the same time it’s pretty unreliable. It may overestimate the energy provided to your body by up to 25 per cent. However, if you use that method to monitor your intake, you can’t do anything about that. And most likely you’ll stick to that anyway.

In their diets, many people rely on CICO, which means ‘calories in, calories out’. It’s an idea that tells us that by calculating your energy intake and expenditure, you can keep yourself in a calorie deficit or surplus. And thus lose or gain weight, respectively. However, not everything is as simple as that. To make it as simple as possible, let’s focus on basics — calories counting.

How calories are calculated?

Calories are counting in a device called ‘calorimeter’. A portion of food is placed into it and burned. Then, researchers see how much energy it contains. The heat is absorbed by water. One calorie is the amount of energy needed to raise the temperature of one gram of water by one degree Celsius. As these amounts are relatively small, experts usually use kilocalories, kcal. One kilocalorie (kcal) equals 1,000 calories. And those are the amounts you see on food labels.

The are many different calorimeters. One of those used by food manufacturers looks like in the picture below.

By Harbor1 – Own work, CC BY 3.0,

Still, it seems to be pretty easy so far, right? So wait for it…

Wilbur Olin Atwater

This calorie-counting system was created in the late 1800s by Wilbur Olin Atwater[1], an American chemist known for his studies of human nutrition and metabolism. He was a scientist at the US Department of Agriculture. Anyway, it was modified over the past 100 years, but it doesn’t mean that the final numbers we can see on food labels are accurate. Also, there are a couple of other aspects we need to take into consideration to have a full(er) picture of calories counting.

How accurate food labels are?

There are some specific regulations within the EU[2]. They include the uncertainty of measurements associated with a measured value. Here is the table telling us what tolerances for foods (other than food supplements) including measurement uncertainty are.

All right. Loads of numbers. By way of example let’s have a look at a real-life situation. In other words, let’s see how much of sugar may contain a product with a nutrition declaration of sugar of 8.5g per 100g of product.

According to rounding guidelines, this equals 8.45 to 8.54g of sugar per 100g of product.

Lower tolerance: lower value (8.45) minus lower tolerance for sugar which is 2g. 8.45 – 2 = 6.45g / 100g. So according to the rounding guidelines, the lower bound tolerance will be 6.5g / 100g.

Upper tolerance: upper value (8.54) plus the upper tolerance for sugar which is 2g. 8.54 + 2 = 10.54g / 100g. So according to the rounding guidelines, the upper bound tolerance will be 11g / 100g.

Furthermore, if the official control finds a sugar content within the range of 6.5 to 11g / 100g, this product is found to be within the tolerance range according to the criteria.

To sum up, if you can see that a specific product, according to its food label, contains 8.5g of sugar per 100g, it may vary between 6.45g and 11g per 100g. It a pretty largeish discrepancy, isn’t it? And we are still talking about sugar content alone, not to mention fat, carbohydrates, protein fibre and salt content. Of course, the tolerance for any of them varies. However, they are still there, affecting food labels information.

How much calories human body can absorb?

You may be overwhelmed already how inaccurate calories counting is. But that’s not the end, unfortunately. There’s another side of the coin. The human body. It’s a pretty complicated ‘machine’ with many, many dependencies and processes within. As a consequence, it needs some energy to make it possible and to keep you alive, you deliver it with food. However, the food provided doesn’t equal the food absorbed. All of that, because your body needs to digest it first and it needs some energy too! Therefore, the amount of calories absorbed by your body is lower than you deliver to it, i.e. you ate.

In other words, digesting food costs energy. And it doesn’t start in your stomach. Eating alone costs calories too. Your body needs calories to chew, swallow, churn, make the acid in the stomach, make the enzymes, to make the rhythmic muscular contractions known as peristalsis that drive the food through, and so forth.

On average, a person uses about 10% of their daily energy expenditure digesting and absorbing food. However, this percentage changes depending on the type of food you eat.

Protein takes the most energy to digest. It’s 20-30% of total calories in protein eaten go to digesting it. For example, if you eat 100 kcal from protein (25g of protein), your body uses 20-30 of those calories to digest and absorb the protein. You’d be then left with 70-80 calories (17.5-20g of protein).

Next is carbohydrates, that takes 5-10% of total calories from it to digest them. So if you eat 100 kcal from carbs (20-25g of carbs), your body uses 90-95 calories out of it (18.5-19g of carbs).

The most effective, unfortunately, is fat. It takes only 0-3% of total calories from it to digest it. So if you eat 100 kcal of fat (11g of fat), you may end up absorbing 90 to 100 kcal out of it, nearly all of them.[3]

Do you still think ‘a calorie is a calorie’ then?

Differences in food digesting.

Oh dear, it’s getting even weirder than you thought, right? Because you see, not every food is being digested in the same way. Therefore, your body digests some food better than other. That is to say, some foods are not fully digested. Significant portions of some of them are excreted. If so, these calories should not be counted either. Meat and nuts are harder to break down, so your body expends even more energy trying to digest them.

For nuts, estimates of the calories they contain may be as much as 25% too high, according to recent research by David Baer, a nutrition scientist at the US Department of Agriculture. There are also some calories issues with almonds. While they are routinely listed as having about 160 kcal per serving, the real figure is about 120 kcal, says Karen Lapsley, the chief scientist at the California Almond Board.

As you can see, there is a pretty big mess when it comes to calories declarations. So here’s where the natural question arises: should we do anything about that? I mean, should we make any corrections to calorie declaration values? We could, but it’s not as simple as you may think.

So let’s assume we made some corrections and lowered these values. Everything seems ok, right? But there’s another issue. Some nutritionists are concerned that lowering the calorie estimates might send the wrong message. The message, that customers can eat more. And that message could be sent at a time when 62% of adults in England were classified as overweight (a body mass index, BMI, of 25 or above) or obese[4].


As you can see, there are many variables when it comes to calorie counting. They come out from the methodology of measure, uncertainty of measurements, digesting process and absorption capabilities. And most likely couple more of them. However, you may think ‘if counting calories are so unreliable, then what should I do now?’. Or ‘should I still count my calories?’. Or even ‘is it worth doing that at all?’. And so on. And I can understand all of these doubts, seriously. You are not the only one with them in mind. And here’s the best answer to all your questions regarding this.

If you count calories, then keep calm and carry on. As this system is not too precise – or I would even say it’s a bit far from being so – it’s still some kind of a system anyway. Eventually, it has its basis, it has its methodology and, generally, it is working.

Eventually, if you measured people height only in feet alone, instead of feet and inches, it would still be some kind of measurement. It wouldn’t tell you the exact height of two people, but you could easily tell looking at their measurements, like 5 feet and 7 feet, that one person is definitely taller than other. Hence, it’s working at some level.

All right, I exaggerated here a bit. But I did that intentionally, to show you the bigger picture. Some measurements may be not too accurate, but they still are some kind of a system, they are standardised and more or less (hopefully more!) just do the job. Like the one with calorie counting.

Of course, it could be more precise. But before we make any changes, we should ask if it’s necessary at all? Is it worth it at all? And will it be reasonably easy applicable to make use of it? I will ask these (and more) questions in my next article.